Key Female Characters in Film Have More to Talk About Besides Men: Automating the Bechdel Test

Human Language Technologies: The 2015 Annual Conference of the North American Chapter of the ACL, pages 830–840, Denver, Colorado, May 31 – June 5, 2015. c⃝2015 Association for Computational Linguistics Key Female Characters in Film Have More to Talk About Besides Men: Automating the Bechdel Test Apoorv Agarwal Computer Science Department Columbia University NY, USA apoorv@cs.columbia.edu Jiehan Zheng† Trinity College of Arts and Sciences Duke University NC, USA jiehan.zheng@duke.edu Shruti Vasanth Kamath† Columbia University NY, USA svk2113@columbia.edu Sriram Balasubramanian† Facebook CA, USA grambler@fb.com Shirin Ann Dey Columbia University NY, USA sad2166@columbia.edu Abstract The Bechdel test is a sequence of three questions designed to assess the presence of women in movies. Many believe that be- cause women are seldom represented in film as strong leaders and thinkers, viewers asso- ciate weaker stereotypes with women. In this paper, we present a computational approach to automate the task of finding whether a movie passes or fails the Bechdel test. This allows us to study the key differences in language use and in the importance of roles of women in movies that pass the test versus the movies that fail the test. Our experiments confirm that in movies that fail the test, women are in fact por- trayed as less-central and less-important char- acters. 1 Introduction The Bechdel test is a series of three questions, which originated from Alison Bechdel’s comic “Dykes to Watch Out For” (Bechdel, 1986). The three ques- tions (or tests) are as follows: (T1) are there at least two named women in the movie? (T2) do these women talk to each other? and (T3) do these women talk to each other about something besides a man? If after watching a movie, the viewers answer “yes” to all three questions, that movie is said to pass the Bechdel test. †These authors contributed equally. The work was done while Jiehan Zheng was at Peddie School and Sriram Balasub- ramanian was at Columbia University. The test was designed to assess the presence of women in movies. Some researchers have embraced the test as an effective primary detector for male bias (Scheiner-Fisher and Russell III, 2012). Due to its generality, the Bechdel test has also been used to as- sess the presence of women in dialogues held on so- cial media platforms such as MySpace and Twitter (Garcia et al., 2014). Several researchers have noted that gender inequality roots itself in both the subcon- scious of individuals and the culture of society as a whole (ˇZiˇzek, 1989; Michel et al., 2011; Garc´ıa and Tanase, 2013). Therefore, combining the Bechdel test with computational analysis can allow for the exposure of gender inequality over a large body of films and literature, thus having the potential to alert society of the necessity to challenge the status quo of male dominance. In this paper, we investigate the task of automat- ing the Bechdel test. In doing so, we aim to study the effectiveness of various linguistic and social net- work analysis features developed for conducting this task. Our …

Content truncated to 500 words